RESEARCH PROFESSIONAL · FORT VALLEY STATE UNIVERSITY
Aftab Siddique, Ph.D. Connecting AI, animal health, and food systems through research grounded in biological expertise.
01 / THE SERIOUS SCIENTIST
Selected research, with study context and validation limits made explicit.
Machine learning and multi-frequency impedance for caseous lymphadenitis screening in goats.
36 goats; 2,673 measurements. Five-fold ROC-AUC: 0.851. Animal-held-out LOSO ROC-AUC: 0.68. Research-stage screening evidence requiring external validation.
Exploring anemia signatures through prepared blood-spot images in goats, sheep, and cattle.
Image-level performance varies across species. Prospective field validation is needed before diagnostic deployment.
Time-aware learning for behavioral anomaly monitoring and short-horizon forecasting in goats.
30 Boer goats; 85,890 RFID records. Five-minute onset ROC-AUC: 0.869; ten-minute: 0.841. Behavioral forecasting does not establish disease diagnosis.
A background spanning biotechnology, animal science, food science, and applied AI.
Source inventory: 40 papers/proceedings, 12 chapters, and 63 conference records. These categories include different contribution types; individual source notes and publication status are retained.
Teaching contributions, research training, and graduate committee mentorship.
Roles and funding status are shown individually. Entries awaiting confirmation are not presented as confirmed awards.
A curated source archive, not an exhaustive record of every online mention.
02 / THE CREATIVE PERSONALITY
My background moves from biotechnology to animal science and food science. That range shapes how I approach sensing, data, and biological problems.
I’m interested in the unexpected connection: what a photograph, an electrical measurement, or a change in activity might help us understand.
Electrical measurements, images, and behavior offer different views of the same biological world.
Animal science asks the question. Food science adds perspective. AI helps explore the patterns.
Interpretation matters as much as prediction.
03 / THE BOLD INNOVATOR
The next step is to connect measurements, models, and field realities—without losing sight of what still needs to be tested.
Explore how complementary sensing modalities can describe animal health more fully than one measurement alone.
Prioritize held-out animals, new settings, interpretable models, and clear limits on what results establish.
Work toward accessible sensing and decision support that respects biological expertise and real operating conditions.
These are research priorities and future directions, not claims of validated commercial products.